Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(6) — a bounded solution of (1) keeps g_n(i) within a constant of ng + v_i

Proved
DermanDenumerable.AvgCost.eq6_value_iteration_bound

by mikedeng1 · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

average-costdynamic-programmingp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1value-iteration

Under conditions (A) and (B), suppose a number ggg and a bounded set of numbers {vj}j∈I\{v_j\}_{j\in I}{vj​}j∈I​ satisfy the optimality equation (1),

g+vi=min⁡k{wik+∑j∈Iqij(k)vj},i∈I.g + v_i = \min_k \Big\{ w_{ik} + \sum_{j\in I} q_{ij}(k) v_j \Big\}, \qquad i \in I.g+vi​=kmin​{wik​+j∈I∑​qij​(k)vj​},i∈I.

Let gn(i)g_n(i)gn​(i) be the value iteration of (5): g0(i)=min⁡kwikg_0(i) = \min_k w_{ik}g0​(i)=mink​wik​ and gn+1(i)=min⁡k{wik+∑jqij(k)gn(j)}g_{n+1}(i) = \min_k \{ w_{ik} + \sum_{j} q_{ij}(k) g_n(j)\}gn+1​(i)=mink​{wik​+∑j​qij​(k)gn​(j)}. Then there is a constant CCC such that

ng+vi−C≤gn(i)≤ng+vi+C,i∈I, n=0,1,2,…n g + v_i - C \le g_n(i) \le n g + v_i + C, \qquad i \in I,\ n = 0, 1, 2, \dotsng+vi​−C≤gn​(i)≤ng+vi​+C,i∈I, n=0,1,2,…

This is display (6) of the paper. Together with the finite-horizon optimality of gng_ngn​ it yields the lower bound g≤QR(i)g \le Q_R(i)g≤QR​(i) for every rule in the proof of Theorem 1.

Formalization Note. The paper calls the constant MMM; it is renamed CCC because M names the model. The series ∑jqij(k)gn(j)\sum_j q_{ij}(k) g_n(j)∑j​qij​(k)gn​(j) converges absolutely since ∣gn∣≤(n+1)sup⁡∣w∣|g_n| \le (n+1)\sup|w|∣gn​∣≤(n+1)sup∣w∣.

Preamble
import Mathlib
import Definitions.Def_DermanDenumerable_AvgCost_Model
open scoped Topology
open Filter SennottDP.AvgFinite
Formal statement
namespace DermanDenumerable.AvgCost

/-- Derman (1966), §3, display (6), p. 1549: if `{g, v_j}` with `v` bounded solves (1), there is a
constant `C` (the paper's `M`) with `n g + v_i − C ≤ g_n(i) ≤ n g + v_i + C` for all `n` and `i`,
where `g_n` is defined by (5). -/
theorem eq6_value_iteration_bound {S Act : Type} [Countable S] (M : MDC S Act)
    (w : S → Act → ℝ) (hB : CostBounded M w) (g : ℝ) (v : S → ℝ) (hv : BddFun v)
    (h1 : IsACOE M w g v) :
    ∃ C : ℝ, ∀ (n : ℕ) (i : S),
      (n : ℝ) * g + v i - C ≤ valueIter M w n i ∧ valueIter M w n i ≤ (n : ℝ) * g + v i + C := by sorry

end DermanDenumerable.AvgCost
Source
Derman, Denumerable State Markovian Decision Processes—Average Cost Criterion, Ann. Math. Statist. 37(6) (1966), p. 1549, §3, displays (5) and (6)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me